From 9a4caf1e9fa4864ce21ba9584a2c336bfbc72740 Mon Sep 17 00:00:00 2001 From: Johannes Weiner Date: Wed, 3 May 2017 14:52:45 -0700 Subject: mm: memcontrol: provide shmem statistics Cgroups currently don't report how much shmem they use, which can be useful data to have, in particular since shmem is included in the cache/file item while being reclaimed like anonymous memory. Add a counter to track shmem pages during charging and uncharging. Link: http://lkml.kernel.org/r/20170221164343.32252-1-hannes@cmpxchg.org Signed-off-by: Johannes Weiner Reported-by: Chris Down Cc: Michal Hocko Cc: Vladimir Davydov Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/memcontrol.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/linux/memcontrol.h') diff --git a/include/linux/memcontrol.h b/include/linux/memcontrol.h index bb7250c45cb8..c5ebb32fef49 100644 --- a/include/linux/memcontrol.h +++ b/include/linux/memcontrol.h @@ -46,6 +46,7 @@ enum mem_cgroup_stat_index { MEM_CGROUP_STAT_CACHE, /* # of pages charged as cache */ MEM_CGROUP_STAT_RSS, /* # of pages charged as anon rss */ MEM_CGROUP_STAT_RSS_HUGE, /* # of pages charged as anon huge */ + MEM_CGROUP_STAT_SHMEM, /* # of pages charged as shmem */ MEM_CGROUP_STAT_FILE_MAPPED, /* # of pages charged as file rss */ MEM_CGROUP_STAT_DIRTY, /* # of dirty pages in page cache */ MEM_CGROUP_STAT_WRITEBACK, /* # of pages under writeback */ -- cgit v1.2.3 From 2a2e48854d704214dac7546e87ae0e4daa0e61a0 Mon Sep 17 00:00:00 2001 From: Johannes Weiner Date: Wed, 3 May 2017 14:55:03 -0700 Subject: mm: vmscan: fix IO/refault regression in cache workingset transition Since commit 59dc76b0d4df ("mm: vmscan: reduce size of inactive file list") we noticed bigger IO spikes during changes in cache access patterns. The patch in question shrunk the inactive list size to leave more room for the current workingset in the presence of streaming IO. However, workingset transitions that previously happened on the inactive list are now pushed out of memory and incur more refaults to complete. This patch disables active list protection when refaults are being observed. This accelerates workingset transitions, and allows more of the new set to establish itself from memory, without eating into the ability to protect the established workingset during stable periods. The workloads that were measurably affected for us were hit pretty bad by it, with refault/majfault rates doubling and tripling during cache transitions, and the machines sustaining half-hour periods of 100% IO utilization, where they'd previously have sub-minute peaks at 60-90%. Stateful services that handle user data tend to be more conservative with kernel upgrades. As a result we hit most page cache issues with some delay, as was the case here. The severity seemed to warrant a stable tag. Fixes: 59dc76b0d4df ("mm: vmscan: reduce size of inactive file list") Link: http://lkml.kernel.org/r/20170404220052.27593-1-hannes@cmpxchg.org Signed-off-by: Johannes Weiner Cc: Rik van Riel Cc: Mel Gorman Cc: Michal Hocko Cc: Vladimir Davydov Cc: [4.7+] Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/memcontrol.h | 64 +++++++++++++++++++++++++++++-- include/linux/mmzone.h | 2 + mm/memcontrol.c | 24 ++++-------- mm/vmscan.c | 94 ++++++++++++++++++++++++++++++++++++---------- mm/workingset.c | 7 +++- 5 files changed, 150 insertions(+), 41 deletions(-) (limited to 'include/linux/memcontrol.h') diff --git a/include/linux/memcontrol.h b/include/linux/memcontrol.h index c5ebb32fef49..cfa91a3ca0ca 100644 --- a/include/linux/memcontrol.h +++ b/include/linux/memcontrol.h @@ -57,6 +57,9 @@ enum mem_cgroup_stat_index { MEMCG_SLAB_RECLAIMABLE, MEMCG_SLAB_UNRECLAIMABLE, MEMCG_SOCK, + MEMCG_WORKINGSET_REFAULT, + MEMCG_WORKINGSET_ACTIVATE, + MEMCG_WORKINGSET_NODERECLAIM, MEMCG_NR_STAT, }; @@ -495,6 +498,40 @@ extern int do_swap_account; void lock_page_memcg(struct page *page); void unlock_page_memcg(struct page *page); +static inline unsigned long mem_cgroup_read_stat(struct mem_cgroup *memcg, + enum mem_cgroup_stat_index idx) +{ + long val = 0; + int cpu; + + for_each_possible_cpu(cpu) + val += per_cpu(memcg->stat->count[idx], cpu); + + if (val < 0) + val = 0; + + return val; +} + +static inline void mem_cgroup_update_stat(struct mem_cgroup *memcg, + enum mem_cgroup_stat_index idx, int val) +{ + if (!mem_cgroup_disabled()) + this_cpu_add(memcg->stat->count[idx], val); +} + +static inline void mem_cgroup_inc_stat(struct mem_cgroup *memcg, + enum mem_cgroup_stat_index idx) +{ + mem_cgroup_update_stat(memcg, idx, 1); +} + +static inline void mem_cgroup_dec_stat(struct mem_cgroup *memcg, + enum mem_cgroup_stat_index idx) +{ + mem_cgroup_update_stat(memcg, idx, -1); +} + /** * mem_cgroup_update_page_stat - update page state statistics * @page: the page @@ -509,14 +546,14 @@ void unlock_page_memcg(struct page *page); * if (TestClearPageState(page)) * mem_cgroup_update_page_stat(page, state, -1); * unlock_page(page) or unlock_page_memcg(page) + * + * Kernel pages are an exception to this, since they'll never move. */ static inline void mem_cgroup_update_page_stat(struct page *page, enum mem_cgroup_stat_index idx, int val) { - VM_BUG_ON(!(rcu_read_lock_held() || PageLocked(page))); - if (page->mem_cgroup) - this_cpu_add(page->mem_cgroup->stat->count[idx], val); + mem_cgroup_update_stat(page->mem_cgroup, idx, val); } static inline void mem_cgroup_inc_page_stat(struct page *page, @@ -741,6 +778,27 @@ static inline bool mem_cgroup_oom_synchronize(bool wait) return false; } +static inline unsigned long mem_cgroup_read_stat(struct mem_cgroup *memcg, + enum mem_cgroup_stat_index idx) +{ + return 0; +} + +static inline void mem_cgroup_update_stat(struct mem_cgroup *memcg, + enum mem_cgroup_stat_index idx, int val) +{ +} + +static inline void mem_cgroup_inc_stat(struct mem_cgroup *memcg, + enum mem_cgroup_stat_index idx) +{ +} + +static inline void mem_cgroup_dec_stat(struct mem_cgroup *memcg, + enum mem_cgroup_stat_index idx) +{ +} + static inline void mem_cgroup_update_page_stat(struct page *page, enum mem_cgroup_stat_index idx, int nr) diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h index 446cf68c1c09..e0c3c5e3d8a0 100644 --- a/include/linux/mmzone.h +++ b/include/linux/mmzone.h @@ -225,6 +225,8 @@ struct lruvec { struct zone_reclaim_stat reclaim_stat; /* Evictions & activations on the inactive file list */ atomic_long_t inactive_age; + /* Refaults at the time of last reclaim cycle */ + unsigned long refaults; #ifdef CONFIG_MEMCG struct pglist_data *pgdat; #endif diff --git a/mm/memcontrol.c b/mm/memcontrol.c index 490d5b4676c1..108d5b097db1 100644 --- a/mm/memcontrol.c +++ b/mm/memcontrol.c @@ -569,23 +569,6 @@ mem_cgroup_largest_soft_limit_node(struct mem_cgroup_tree_per_node *mctz) * common workload, threshold and synchronization as vmstat[] should be * implemented. */ -static unsigned long -mem_cgroup_read_stat(struct mem_cgroup *memcg, enum mem_cgroup_stat_index idx) -{ - long val = 0; - int cpu; - - /* Per-cpu values can be negative, use a signed accumulator */ - for_each_possible_cpu(cpu) - val += per_cpu(memcg->stat->count[idx], cpu); - /* - * Summing races with updates, so val may be negative. Avoid exposing - * transient negative values. - */ - if (val < 0) - val = 0; - return val; -} static unsigned long mem_cgroup_read_events(struct mem_cgroup *memcg, enum mem_cgroup_events_index idx) @@ -5244,6 +5227,13 @@ static int memory_stat_show(struct seq_file *m, void *v) seq_printf(m, "pgmajfault %lu\n", events[MEM_CGROUP_EVENTS_PGMAJFAULT]); + seq_printf(m, "workingset_refault %lu\n", + stat[MEMCG_WORKINGSET_REFAULT]); + seq_printf(m, "workingset_activate %lu\n", + stat[MEMCG_WORKINGSET_ACTIVATE]); + seq_printf(m, "workingset_nodereclaim %lu\n", + stat[MEMCG_WORKINGSET_NODERECLAIM]); + return 0; } diff --git a/mm/vmscan.c b/mm/vmscan.c index 7a30150b4dee..e8cb983a8c84 100644 --- a/mm/vmscan.c +++ b/mm/vmscan.c @@ -2006,6 +2006,8 @@ static void shrink_active_list(unsigned long nr_to_scan, * Both inactive lists should also be large enough that each inactive * page has a chance to be referenced again before it is reclaimed. * + * If that fails and refaulting is observed, the inactive list grows. + * * The inactive_ratio is the target ratio of ACTIVE to INACTIVE pages * on this LRU, maintained by the pageout code. A zone->inactive_ratio * of 3 means 3:1 or 25% of the pages are kept on the inactive list. @@ -2022,12 +2024,15 @@ static void shrink_active_list(unsigned long nr_to_scan, * 10TB 320 32GB */ static bool inactive_list_is_low(struct lruvec *lruvec, bool file, - struct scan_control *sc, bool trace) + struct mem_cgroup *memcg, + struct scan_control *sc, bool actual_reclaim) { - unsigned long inactive_ratio; - unsigned long inactive, active; - enum lru_list inactive_lru = file * LRU_FILE; enum lru_list active_lru = file * LRU_FILE + LRU_ACTIVE; + struct pglist_data *pgdat = lruvec_pgdat(lruvec); + enum lru_list inactive_lru = file * LRU_FILE; + unsigned long inactive, active; + unsigned long inactive_ratio; + unsigned long refaults; unsigned long gb; /* @@ -2040,27 +2045,43 @@ static bool inactive_list_is_low(struct lruvec *lruvec, bool file, inactive = lruvec_lru_size(lruvec, inactive_lru, sc->reclaim_idx); active = lruvec_lru_size(lruvec, active_lru, sc->reclaim_idx); - gb = (inactive + active) >> (30 - PAGE_SHIFT); - if (gb) - inactive_ratio = int_sqrt(10 * gb); + if (memcg) + refaults = mem_cgroup_read_stat(memcg, + MEMCG_WORKINGSET_ACTIVATE); else - inactive_ratio = 1; + refaults = node_page_state(pgdat, WORKINGSET_ACTIVATE); + + /* + * When refaults are being observed, it means a new workingset + * is being established. Disable active list protection to get + * rid of the stale workingset quickly. + */ + if (file && actual_reclaim && lruvec->refaults != refaults) { + inactive_ratio = 0; + } else { + gb = (inactive + active) >> (30 - PAGE_SHIFT); + if (gb) + inactive_ratio = int_sqrt(10 * gb); + else + inactive_ratio = 1; + } - if (trace) - trace_mm_vmscan_inactive_list_is_low(lruvec_pgdat(lruvec)->node_id, - sc->reclaim_idx, - lruvec_lru_size(lruvec, inactive_lru, MAX_NR_ZONES), inactive, - lruvec_lru_size(lruvec, active_lru, MAX_NR_ZONES), active, - inactive_ratio, file); + if (actual_reclaim) + trace_mm_vmscan_inactive_list_is_low(pgdat->node_id, sc->reclaim_idx, + lruvec_lru_size(lruvec, inactive_lru, MAX_NR_ZONES), inactive, + lruvec_lru_size(lruvec, active_lru, MAX_NR_ZONES), active, + inactive_ratio, file); return inactive * inactive_ratio < active; } static unsigned long shrink_list(enum lru_list lru, unsigned long nr_to_scan, - struct lruvec *lruvec, struct scan_control *sc) + struct lruvec *lruvec, struct mem_cgroup *memcg, + struct scan_control *sc) { if (is_active_lru(lru)) { - if (inactive_list_is_low(lruvec, is_file_lru(lru), sc, true)) + if (inactive_list_is_low(lruvec, is_file_lru(lru), + memcg, sc, true)) shrink_active_list(nr_to_scan, lruvec, sc, lru); return 0; } @@ -2169,7 +2190,7 @@ static void get_scan_count(struct lruvec *lruvec, struct mem_cgroup *memcg, * lruvec even if it has plenty of old anonymous pages unless the * system is under heavy pressure. */ - if (!inactive_list_is_low(lruvec, true, sc, false) && + if (!inactive_list_is_low(lruvec, true, memcg, sc, false) && lruvec_lru_size(lruvec, LRU_INACTIVE_FILE, sc->reclaim_idx) >> sc->priority) { scan_balance = SCAN_FILE; goto out; @@ -2320,7 +2341,7 @@ static void shrink_node_memcg(struct pglist_data *pgdat, struct mem_cgroup *memc nr[lru] -= nr_to_scan; nr_reclaimed += shrink_list(lru, nr_to_scan, - lruvec, sc); + lruvec, memcg, sc); } } @@ -2387,7 +2408,7 @@ static void shrink_node_memcg(struct pglist_data *pgdat, struct mem_cgroup *memc * Even if we did not try to evict anon pages at all, we want to * rebalance the anon lru active/inactive ratio. */ - if (inactive_list_is_low(lruvec, false, sc, true)) + if (inactive_list_is_low(lruvec, false, memcg, sc, true)) shrink_active_list(SWAP_CLUSTER_MAX, lruvec, sc, LRU_ACTIVE_ANON); } @@ -2703,6 +2724,26 @@ static void shrink_zones(struct zonelist *zonelist, struct scan_control *sc) sc->gfp_mask = orig_mask; } +static void snapshot_refaults(struct mem_cgroup *root_memcg, pg_data_t *pgdat) +{ + struct mem_cgroup *memcg; + + memcg = mem_cgroup_iter(root_memcg, NULL, NULL); + do { + unsigned long refaults; + struct lruvec *lruvec; + + if (memcg) + refaults = mem_cgroup_read_stat(memcg, + MEMCG_WORKINGSET_ACTIVATE); + else + refaults = node_page_state(pgdat, WORKINGSET_ACTIVATE); + + lruvec = mem_cgroup_lruvec(pgdat, memcg); + lruvec->refaults = refaults; + } while ((memcg = mem_cgroup_iter(root_memcg, memcg, NULL))); +} + /* * This is the main entry point to direct page reclaim. * @@ -2723,6 +2764,9 @@ static unsigned long do_try_to_free_pages(struct zonelist *zonelist, struct scan_control *sc) { int initial_priority = sc->priority; + pg_data_t *last_pgdat; + struct zoneref *z; + struct zone *zone; retry: delayacct_freepages_start(); @@ -2749,6 +2793,15 @@ retry: sc->may_writepage = 1; } while (--sc->priority >= 0); + last_pgdat = NULL; + for_each_zone_zonelist_nodemask(zone, z, zonelist, sc->reclaim_idx, + sc->nodemask) { + if (zone->zone_pgdat == last_pgdat) + continue; + last_pgdat = zone->zone_pgdat; + snapshot_refaults(sc->target_mem_cgroup, zone->zone_pgdat); + } + delayacct_freepages_end(); if (sc->nr_reclaimed) @@ -3033,7 +3086,7 @@ static void age_active_anon(struct pglist_data *pgdat, do { struct lruvec *lruvec = mem_cgroup_lruvec(pgdat, memcg); - if (inactive_list_is_low(lruvec, false, sc, true)) + if (inactive_list_is_low(lruvec, false, memcg, sc, true)) shrink_active_list(SWAP_CLUSTER_MAX, lruvec, sc, LRU_ACTIVE_ANON); @@ -3280,6 +3333,7 @@ static int balance_pgdat(pg_data_t *pgdat, int order, int classzone_idx) pgdat->kswapd_failures++; out: + snapshot_refaults(NULL, pgdat); /* * Return the order kswapd stopped reclaiming at as * prepare_kswapd_sleep() takes it into account. If another caller diff --git a/mm/workingset.c b/mm/workingset.c index eda05c71fa49..51c6f61d4cea 100644 --- a/mm/workingset.c +++ b/mm/workingset.c @@ -269,7 +269,6 @@ bool workingset_refault(void *shadow) lruvec = mem_cgroup_lruvec(pgdat, memcg); refault = atomic_long_read(&lruvec->inactive_age); active_file = lruvec_lru_size(lruvec, LRU_ACTIVE_FILE, MAX_NR_ZONES); - rcu_read_unlock(); /* * The unsigned subtraction here gives an accurate distance @@ -290,11 +289,15 @@ bool workingset_refault(void *shadow) refault_distance = (refault - eviction) & EVICTION_MASK; inc_node_state(pgdat, WORKINGSET_REFAULT); + mem_cgroup_inc_stat(memcg, MEMCG_WORKINGSET_REFAULT); if (refault_distance <= active_file) { inc_node_state(pgdat, WORKINGSET_ACTIVATE); + mem_cgroup_inc_stat(memcg, MEMCG_WORKINGSET_ACTIVATE); + rcu_read_unlock(); return true; } + rcu_read_unlock(); return false; } @@ -472,6 +475,8 @@ static enum lru_status shadow_lru_isolate(struct list_head *item, if (WARN_ON_ONCE(node->exceptional)) goto out_invalid; inc_node_state(page_pgdat(virt_to_page(node)), WORKINGSET_NODERECLAIM); + mem_cgroup_inc_page_stat(virt_to_page(node), + MEMCG_WORKINGSET_NODERECLAIM); __radix_tree_delete_node(&mapping->page_tree, node, workingset_update_node, mapping); -- cgit v1.2.3 From 31176c781508e4e35b1cc4ae2f0a5abd1f4ea689 Mon Sep 17 00:00:00 2001 From: Johannes Weiner Date: Wed, 3 May 2017 14:55:07 -0700 Subject: mm: memcontrol: clean up memory.events counting function We only ever count single events, drop the @nr parameter. Rename the function accordingly. Remove low-information kerneldoc. Link: http://lkml.kernel.org/r/20170404220148.28338-1-hannes@cmpxchg.org Signed-off-by: Johannes Weiner Acked-by: Vladimir Davydov Acked-by: Michal Hocko Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/memcontrol.h | 18 +++++------------- mm/memcontrol.c | 8 ++++---- mm/vmscan.c | 2 +- 3 files changed, 10 insertions(+), 18 deletions(-) (limited to 'include/linux/memcontrol.h') diff --git a/include/linux/memcontrol.h b/include/linux/memcontrol.h index cfa91a3ca0ca..bc0c16e284c0 100644 --- a/include/linux/memcontrol.h +++ b/include/linux/memcontrol.h @@ -287,17 +287,10 @@ static inline bool mem_cgroup_disabled(void) return !cgroup_subsys_enabled(memory_cgrp_subsys); } -/** - * mem_cgroup_events - count memory events against a cgroup - * @memcg: the memory cgroup - * @idx: the event index - * @nr: the number of events to account for - */ -static inline void mem_cgroup_events(struct mem_cgroup *memcg, - enum mem_cgroup_events_index idx, - unsigned int nr) +static inline void mem_cgroup_event(struct mem_cgroup *memcg, + enum mem_cgroup_events_index idx) { - this_cpu_add(memcg->stat->events[idx], nr); + this_cpu_inc(memcg->stat->events[idx]); cgroup_file_notify(&memcg->events_file); } @@ -614,9 +607,8 @@ static inline bool mem_cgroup_disabled(void) return true; } -static inline void mem_cgroup_events(struct mem_cgroup *memcg, - enum mem_cgroup_events_index idx, - unsigned int nr) +static inline void mem_cgroup_event(struct mem_cgroup *memcg, + enum mem_cgroup_events_index idx) { } diff --git a/mm/memcontrol.c b/mm/memcontrol.c index 108d5b097db1..1ffa3ad201ea 100644 --- a/mm/memcontrol.c +++ b/mm/memcontrol.c @@ -1825,7 +1825,7 @@ static void reclaim_high(struct mem_cgroup *memcg, do { if (page_counter_read(&memcg->memory) <= memcg->high) continue; - mem_cgroup_events(memcg, MEMCG_HIGH, 1); + mem_cgroup_event(memcg, MEMCG_HIGH); try_to_free_mem_cgroup_pages(memcg, nr_pages, gfp_mask, true); } while ((memcg = parent_mem_cgroup(memcg))); } @@ -1916,7 +1916,7 @@ retry: if (!gfpflags_allow_blocking(gfp_mask)) goto nomem; - mem_cgroup_events(mem_over_limit, MEMCG_MAX, 1); + mem_cgroup_event(mem_over_limit, MEMCG_MAX); nr_reclaimed = try_to_free_mem_cgroup_pages(mem_over_limit, nr_pages, gfp_mask, may_swap); @@ -1959,7 +1959,7 @@ retry: if (fatal_signal_pending(current)) goto force; - mem_cgroup_events(mem_over_limit, MEMCG_OOM, 1); + mem_cgroup_event(mem_over_limit, MEMCG_OOM); mem_cgroup_oom(mem_over_limit, gfp_mask, get_order(nr_pages * PAGE_SIZE)); @@ -5142,7 +5142,7 @@ static ssize_t memory_max_write(struct kernfs_open_file *of, continue; } - mem_cgroup_events(memcg, MEMCG_OOM, 1); + mem_cgroup_event(memcg, MEMCG_OOM); if (!mem_cgroup_out_of_memory(memcg, GFP_KERNEL, 0)) break; } diff --git a/mm/vmscan.c b/mm/vmscan.c index e8cb983a8c84..fbec74af2b69 100644 --- a/mm/vmscan.c +++ b/mm/vmscan.c @@ -2526,7 +2526,7 @@ static bool shrink_node(pg_data_t *pgdat, struct scan_control *sc) sc->memcg_low_skipped = 1; continue; } - mem_cgroup_events(memcg, MEMCG_LOW, 1); + mem_cgroup_event(memcg, MEMCG_LOW); } reclaimed = sc->nr_reclaimed; -- cgit v1.2.3 From df0e53d0619e83b465e363c088bf4eeb2848273b Mon Sep 17 00:00:00 2001 From: Johannes Weiner Date: Wed, 3 May 2017 14:55:10 -0700 Subject: mm: memcontrol: re-use global VM event enum The current duplication is a high-maintenance mess, and it's painful to add new items. This increases the size of the event array, but we'll eventually want most of the VM events tracked on a per-cgroup basis anyway. Link: http://lkml.kernel.org/r/20170404220148.28338-2-hannes@cmpxchg.org Signed-off-by: Johannes Weiner Acked-by: Vladimir Davydov Cc: Michal Hocko Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/memcontrol.h | 45 ++++++++++++--------------------------- mm/memcontrol.c | 53 ++++++++++++++++++++++++---------------------- 2 files changed, 42 insertions(+), 56 deletions(-) (limited to 'include/linux/memcontrol.h') diff --git a/include/linux/memcontrol.h b/include/linux/memcontrol.h index bc0c16e284c0..0bb5f055bd26 100644 --- a/include/linux/memcontrol.h +++ b/include/linux/memcontrol.h @@ -69,20 +69,6 @@ struct mem_cgroup_reclaim_cookie { unsigned int generation; }; -enum mem_cgroup_events_index { - MEM_CGROUP_EVENTS_PGPGIN, /* # of pages paged in */ - MEM_CGROUP_EVENTS_PGPGOUT, /* # of pages paged out */ - MEM_CGROUP_EVENTS_PGFAULT, /* # of page-faults */ - MEM_CGROUP_EVENTS_PGMAJFAULT, /* # of major page-faults */ - MEM_CGROUP_EVENTS_NSTATS, - /* default hierarchy events */ - MEMCG_LOW = MEM_CGROUP_EVENTS_NSTATS, - MEMCG_HIGH, - MEMCG_MAX, - MEMCG_OOM, - MEMCG_NR_EVENTS, -}; - /* * Per memcg event counter is incremented at every pagein/pageout. With THP, * it will be incremated by the number of pages. This counter is used for @@ -106,6 +92,15 @@ struct mem_cgroup_id { atomic_t ref; }; +/* Cgroup-specific events, on top of universal VM events */ +enum memcg_event_item { + MEMCG_LOW = NR_VM_EVENT_ITEMS, + MEMCG_HIGH, + MEMCG_MAX, + MEMCG_OOM, + MEMCG_NR_EVENTS, +}; + struct mem_cgroup_stat_cpu { long count[MEMCG_NR_STAT]; unsigned long events[MEMCG_NR_EVENTS]; @@ -288,9 +283,9 @@ static inline bool mem_cgroup_disabled(void) } static inline void mem_cgroup_event(struct mem_cgroup *memcg, - enum mem_cgroup_events_index idx) + enum memcg_event_item event) { - this_cpu_inc(memcg->stat->events[idx]); + this_cpu_inc(memcg->stat->events[event]); cgroup_file_notify(&memcg->events_file); } @@ -575,20 +570,8 @@ static inline void mem_cgroup_count_vm_event(struct mm_struct *mm, rcu_read_lock(); memcg = mem_cgroup_from_task(rcu_dereference(mm->owner)); - if (unlikely(!memcg)) - goto out; - - switch (idx) { - case PGFAULT: - this_cpu_inc(memcg->stat->events[MEM_CGROUP_EVENTS_PGFAULT]); - break; - case PGMAJFAULT: - this_cpu_inc(memcg->stat->events[MEM_CGROUP_EVENTS_PGMAJFAULT]); - break; - default: - BUG(); - } -out: + if (likely(memcg)) + this_cpu_inc(memcg->stat->events[idx]); rcu_read_unlock(); } #ifdef CONFIG_TRANSPARENT_HUGEPAGE @@ -608,7 +591,7 @@ static inline bool mem_cgroup_disabled(void) } static inline void mem_cgroup_event(struct mem_cgroup *memcg, - enum mem_cgroup_events_index idx) + enum memcg_event_item event) { } diff --git a/mm/memcontrol.c b/mm/memcontrol.c index 1ffa3ad201ea..6b42887e5f14 100644 --- a/mm/memcontrol.c +++ b/mm/memcontrol.c @@ -111,13 +111,6 @@ static const char * const mem_cgroup_stat_names[] = { "swap", }; -static const char * const mem_cgroup_events_names[] = { - "pgpgin", - "pgpgout", - "pgfault", - "pgmajfault", -}; - static const char * const mem_cgroup_lru_names[] = { "inactive_anon", "active_anon", @@ -571,13 +564,13 @@ mem_cgroup_largest_soft_limit_node(struct mem_cgroup_tree_per_node *mctz) */ static unsigned long mem_cgroup_read_events(struct mem_cgroup *memcg, - enum mem_cgroup_events_index idx) + enum memcg_event_item event) { unsigned long val = 0; int cpu; for_each_possible_cpu(cpu) - val += per_cpu(memcg->stat->events[idx], cpu); + val += per_cpu(memcg->stat->events[event], cpu); return val; } @@ -608,9 +601,9 @@ static void mem_cgroup_charge_statistics(struct mem_cgroup *memcg, /* pagein of a big page is an event. So, ignore page size */ if (nr_pages > 0) - __this_cpu_inc(memcg->stat->events[MEM_CGROUP_EVENTS_PGPGIN]); + __this_cpu_inc(memcg->stat->events[PGPGIN]); else { - __this_cpu_inc(memcg->stat->events[MEM_CGROUP_EVENTS_PGPGOUT]); + __this_cpu_inc(memcg->stat->events[PGPGOUT]); nr_pages = -nr_pages; /* for event */ } @@ -3119,6 +3112,21 @@ static int memcg_numa_stat_show(struct seq_file *m, void *v) } #endif /* CONFIG_NUMA */ +/* Universal VM events cgroup1 shows, original sort order */ +unsigned int memcg1_events[] = { + PGPGIN, + PGPGOUT, + PGFAULT, + PGMAJFAULT, +}; + +static const char *const memcg1_event_names[] = { + "pgpgin", + "pgpgout", + "pgfault", + "pgmajfault", +}; + static int memcg_stat_show(struct seq_file *m, void *v) { struct mem_cgroup *memcg = mem_cgroup_from_css(seq_css(m)); @@ -3128,8 +3136,6 @@ static int memcg_stat_show(struct seq_file *m, void *v) BUILD_BUG_ON(ARRAY_SIZE(mem_cgroup_stat_names) != MEM_CGROUP_STAT_NSTATS); - BUILD_BUG_ON(ARRAY_SIZE(mem_cgroup_events_names) != - MEM_CGROUP_EVENTS_NSTATS); BUILD_BUG_ON(ARRAY_SIZE(mem_cgroup_lru_names) != NR_LRU_LISTS); for (i = 0; i < MEM_CGROUP_STAT_NSTATS; i++) { @@ -3139,9 +3145,9 @@ static int memcg_stat_show(struct seq_file *m, void *v) mem_cgroup_read_stat(memcg, i) * PAGE_SIZE); } - for (i = 0; i < MEM_CGROUP_EVENTS_NSTATS; i++) - seq_printf(m, "%s %lu\n", mem_cgroup_events_names[i], - mem_cgroup_read_events(memcg, i)); + for (i = 0; i < ARRAY_SIZE(memcg1_events); i++) + seq_printf(m, "%s %lu\n", memcg1_event_names[i], + mem_cgroup_read_events(memcg, memcg1_events[i])); for (i = 0; i < NR_LRU_LISTS; i++) seq_printf(m, "%s %lu\n", mem_cgroup_lru_names[i], @@ -3169,13 +3175,12 @@ static int memcg_stat_show(struct seq_file *m, void *v) seq_printf(m, "total_%s %llu\n", mem_cgroup_stat_names[i], val); } - for (i = 0; i < MEM_CGROUP_EVENTS_NSTATS; i++) { + for (i = 0; i < ARRAY_SIZE(memcg1_events); i++) { unsigned long long val = 0; for_each_mem_cgroup_tree(mi, memcg) - val += mem_cgroup_read_events(mi, i); - seq_printf(m, "total_%s %llu\n", - mem_cgroup_events_names[i], val); + val += mem_cgroup_read_events(mi, memcg1_events[i]); + seq_printf(m, "total_%s %llu\n", memcg1_event_names[i], val); } for (i = 0; i < NR_LRU_LISTS; i++) { @@ -5222,10 +5227,8 @@ static int memory_stat_show(struct seq_file *m, void *v) /* Accumulated memory events */ - seq_printf(m, "pgfault %lu\n", - events[MEM_CGROUP_EVENTS_PGFAULT]); - seq_printf(m, "pgmajfault %lu\n", - events[MEM_CGROUP_EVENTS_PGMAJFAULT]); + seq_printf(m, "pgfault %lu\n", events[PGFAULT]); + seq_printf(m, "pgmajfault %lu\n", events[PGMAJFAULT]); seq_printf(m, "workingset_refault %lu\n", stat[MEMCG_WORKINGSET_REFAULT]); @@ -5493,7 +5496,7 @@ static void uncharge_batch(struct mem_cgroup *memcg, unsigned long pgpgout, __this_cpu_sub(memcg->stat->count[MEM_CGROUP_STAT_CACHE], nr_file); __this_cpu_sub(memcg->stat->count[MEM_CGROUP_STAT_RSS_HUGE], nr_huge); __this_cpu_sub(memcg->stat->count[MEM_CGROUP_STAT_SHMEM], nr_shmem); - __this_cpu_add(memcg->stat->events[MEM_CGROUP_EVENTS_PGPGOUT], pgpgout); + __this_cpu_add(memcg->stat->events[PGPGOUT], pgpgout); __this_cpu_add(memcg->stat->nr_page_events, nr_pages); memcg_check_events(memcg, dummy_page); local_irq_restore(flags); -- cgit v1.2.3 From 71cd31135d4cf030a057ed7079a75a40c0a4a796 Mon Sep 17 00:00:00 2001 From: Johannes Weiner Date: Wed, 3 May 2017 14:55:13 -0700 Subject: mm: memcontrol: re-use node VM page state enum The current duplication is a high-maintenance mess, and it's painful to add new items or query memcg state from the rest of the VM. This increases the size of the stat array marginally, but we should aim to track all these stats on a per-cgroup level anyway. Link: http://lkml.kernel.org/r/20170404220148.28338-3-hannes@cmpxchg.org Signed-off-by: Johannes Weiner Acked-by: Vladimir Davydov Cc: Michal Hocko Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/memcontrol.h | 100 +++++++++++++++------------------ mm/memcontrol.c | 135 +++++++++++++++++++++++---------------------- mm/page-writeback.c | 10 ++-- mm/rmap.c | 4 +- mm/vmscan.c | 5 +- mm/workingset.c | 7 +-- 6 files changed, 123 insertions(+), 138 deletions(-) (limited to 'include/linux/memcontrol.h') diff --git a/include/linux/memcontrol.h b/include/linux/memcontrol.h index 0bb5f055bd26..0fa1f5de6841 100644 --- a/include/linux/memcontrol.h +++ b/include/linux/memcontrol.h @@ -35,40 +35,45 @@ struct page; struct mm_struct; struct kmem_cache; -/* - * The corresponding mem_cgroup_stat_names is defined in mm/memcontrol.c, - * These two lists should keep in accord with each other. - */ -enum mem_cgroup_stat_index { - /* - * For MEM_CONTAINER_TYPE_ALL, usage = pagecache + rss. - */ - MEM_CGROUP_STAT_CACHE, /* # of pages charged as cache */ - MEM_CGROUP_STAT_RSS, /* # of pages charged as anon rss */ - MEM_CGROUP_STAT_RSS_HUGE, /* # of pages charged as anon huge */ - MEM_CGROUP_STAT_SHMEM, /* # of pages charged as shmem */ - MEM_CGROUP_STAT_FILE_MAPPED, /* # of pages charged as file rss */ - MEM_CGROUP_STAT_DIRTY, /* # of dirty pages in page cache */ - MEM_CGROUP_STAT_WRITEBACK, /* # of pages under writeback */ - MEM_CGROUP_STAT_SWAP, /* # of pages, swapped out */ - MEM_CGROUP_STAT_NSTATS, - /* default hierarchy stats */ - MEMCG_KERNEL_STACK_KB = MEM_CGROUP_STAT_NSTATS, +/* Cgroup-specific page state, on top of universal node page state */ +enum memcg_stat_item { + MEMCG_CACHE = NR_VM_NODE_STAT_ITEMS, + MEMCG_RSS, + MEMCG_RSS_HUGE, + MEMCG_SWAP, + MEMCG_SOCK, + /* XXX: why are these zone and not node counters? */ + MEMCG_KERNEL_STACK_KB, MEMCG_SLAB_RECLAIMABLE, MEMCG_SLAB_UNRECLAIMABLE, - MEMCG_SOCK, - MEMCG_WORKINGSET_REFAULT, - MEMCG_WORKINGSET_ACTIVATE, - MEMCG_WORKINGSET_NODERECLAIM, MEMCG_NR_STAT, }; +/* Cgroup-specific events, on top of universal VM events */ +enum memcg_event_item { + MEMCG_LOW = NR_VM_EVENT_ITEMS, + MEMCG_HIGH, + MEMCG_MAX, + MEMCG_OOM, + MEMCG_NR_EVENTS, +}; + struct mem_cgroup_reclaim_cookie { pg_data_t *pgdat; int priority; unsigned int generation; }; +#ifdef CONFIG_MEMCG + +#define MEM_CGROUP_ID_SHIFT 16 +#define MEM_CGROUP_ID_MAX USHRT_MAX + +struct mem_cgroup_id { + int id; + atomic_t ref; +}; + /* * Per memcg event counter is incremented at every pagein/pageout. With THP, * it will be incremated by the number of pages. This counter is used for @@ -82,25 +87,6 @@ enum mem_cgroup_events_target { MEM_CGROUP_NTARGETS, }; -#ifdef CONFIG_MEMCG - -#define MEM_CGROUP_ID_SHIFT 16 -#define MEM_CGROUP_ID_MAX USHRT_MAX - -struct mem_cgroup_id { - int id; - atomic_t ref; -}; - -/* Cgroup-specific events, on top of universal VM events */ -enum memcg_event_item { - MEMCG_LOW = NR_VM_EVENT_ITEMS, - MEMCG_HIGH, - MEMCG_MAX, - MEMCG_OOM, - MEMCG_NR_EVENTS, -}; - struct mem_cgroup_stat_cpu { long count[MEMCG_NR_STAT]; unsigned long events[MEMCG_NR_EVENTS]; @@ -487,7 +473,7 @@ void lock_page_memcg(struct page *page); void unlock_page_memcg(struct page *page); static inline unsigned long mem_cgroup_read_stat(struct mem_cgroup *memcg, - enum mem_cgroup_stat_index idx) + enum memcg_stat_item idx) { long val = 0; int cpu; @@ -502,20 +488,20 @@ static inline unsigned long mem_cgroup_read_stat(struct mem_cgroup *memcg, } static inline void mem_cgroup_update_stat(struct mem_cgroup *memcg, - enum mem_cgroup_stat_index idx, int val) + enum memcg_stat_item idx, int val) { if (!mem_cgroup_disabled()) this_cpu_add(memcg->stat->count[idx], val); } static inline void mem_cgroup_inc_stat(struct mem_cgroup *memcg, - enum mem_cgroup_stat_index idx) + enum memcg_stat_item idx) { mem_cgroup_update_stat(memcg, idx, 1); } static inline void mem_cgroup_dec_stat(struct mem_cgroup *memcg, - enum mem_cgroup_stat_index idx) + enum memcg_stat_item idx) { mem_cgroup_update_stat(memcg, idx, -1); } @@ -538,20 +524,20 @@ static inline void mem_cgroup_dec_stat(struct mem_cgroup *memcg, * Kernel pages are an exception to this, since they'll never move. */ static inline void mem_cgroup_update_page_stat(struct page *page, - enum mem_cgroup_stat_index idx, int val) + enum memcg_stat_item idx, int val) { if (page->mem_cgroup) mem_cgroup_update_stat(page->mem_cgroup, idx, val); } static inline void mem_cgroup_inc_page_stat(struct page *page, - enum mem_cgroup_stat_index idx) + enum memcg_stat_item idx) { mem_cgroup_update_page_stat(page, idx, 1); } static inline void mem_cgroup_dec_page_stat(struct page *page, - enum mem_cgroup_stat_index idx) + enum memcg_stat_item idx) { mem_cgroup_update_page_stat(page, idx, -1); } @@ -760,33 +746,33 @@ static inline unsigned long mem_cgroup_read_stat(struct mem_cgroup *memcg, } static inline void mem_cgroup_update_stat(struct mem_cgroup *memcg, - enum mem_cgroup_stat_index idx, int val) + enum memcg_stat_item idx, int val) { } static inline void mem_cgroup_inc_stat(struct mem_cgroup *memcg, - enum mem_cgroup_stat_index idx) + enum memcg_stat_item idx) { } static inline void mem_cgroup_dec_stat(struct mem_cgroup *memcg, - enum mem_cgroup_stat_index idx) + enum memcg_stat_item idx) { } static inline void mem_cgroup_update_page_stat(struct page *page, - enum mem_cgroup_stat_index idx, + enum memcg_stat_item idx, int nr) { } static inline void mem_cgroup_inc_page_stat(struct page *page, - enum mem_cgroup_stat_index idx) + enum memcg_stat_item idx) { } static inline void mem_cgroup_dec_page_stat(struct page *page, - enum mem_cgroup_stat_index idx) + enum memcg_stat_item idx) { } @@ -906,7 +892,7 @@ static inline int memcg_cache_id(struct mem_cgroup *memcg) * @val: number of pages (positive or negative) */ static inline void memcg_kmem_update_page_stat(struct page *page, - enum mem_cgroup_stat_index idx, int val) + enum memcg_stat_item idx, int val) { if (memcg_kmem_enabled() && page->mem_cgroup) this_cpu_add(page->mem_cgroup->stat->count[idx], val); @@ -935,7 +921,7 @@ static inline void memcg_put_cache_ids(void) } static inline void memcg_kmem_update_page_stat(struct page *page, - enum mem_cgroup_stat_index idx, int val) + enum memcg_stat_item idx, int val) { } #endif /* CONFIG_MEMCG && !CONFIG_SLOB */ diff --git a/mm/memcontrol.c b/mm/memcontrol.c index 6b42887e5f14..6fe4c7fafbfc 100644 --- a/mm/memcontrol.c +++ b/mm/memcontrol.c @@ -100,18 +100,7 @@ static bool do_memsw_account(void) return !cgroup_subsys_on_dfl(memory_cgrp_subsys) && do_swap_account; } -static const char * const mem_cgroup_stat_names[] = { - "cache", - "rss", - "rss_huge", - "shmem", - "mapped_file", - "dirty", - "writeback", - "swap", -}; - -static const char * const mem_cgroup_lru_names[] = { +static const char *const mem_cgroup_lru_names[] = { "inactive_anon", "active_anon", "inactive_file", @@ -583,20 +572,16 @@ static void mem_cgroup_charge_statistics(struct mem_cgroup *memcg, * counted as CACHE even if it's on ANON LRU. */ if (PageAnon(page)) - __this_cpu_add(memcg->stat->count[MEM_CGROUP_STAT_RSS], - nr_pages); + __this_cpu_add(memcg->stat->count[MEMCG_RSS], nr_pages); else { - __this_cpu_add(memcg->stat->count[MEM_CGROUP_STAT_CACHE], - nr_pages); + __this_cpu_add(memcg->stat->count[MEMCG_CACHE], nr_pages); if (PageSwapBacked(page)) - __this_cpu_add(memcg->stat->count[MEM_CGROUP_STAT_SHMEM], - nr_pages); + __this_cpu_add(memcg->stat->count[NR_SHMEM], nr_pages); } if (compound) { VM_BUG_ON_PAGE(!PageTransHuge(page), page); - __this_cpu_add(memcg->stat->count[MEM_CGROUP_STAT_RSS_HUGE], - nr_pages); + __this_cpu_add(memcg->stat->count[MEMCG_RSS_HUGE], nr_pages); } /* pagein of a big page is an event. So, ignore page size */ @@ -1125,6 +1110,28 @@ static bool mem_cgroup_wait_acct_move(struct mem_cgroup *memcg) return false; } +unsigned int memcg1_stats[] = { + MEMCG_CACHE, + MEMCG_RSS, + MEMCG_RSS_HUGE, + NR_SHMEM, + NR_FILE_MAPPED, + NR_FILE_DIRTY, + NR_WRITEBACK, + MEMCG_SWAP, +}; + +static const char *const memcg1_stat_names[] = { + "cache", + "rss", + "rss_huge", + "shmem", + "mapped_file", + "dirty", + "writeback", + "swap", +}; + #define K(x) ((x) << (PAGE_SHIFT-10)) /** * mem_cgroup_print_oom_info: Print OOM information relevant to memory controller. @@ -1169,11 +1176,11 @@ void mem_cgroup_print_oom_info(struct mem_cgroup *memcg, struct task_struct *p) pr_cont_cgroup_path(iter->css.cgroup); pr_cont(":"); - for (i = 0; i < MEM_CGROUP_STAT_NSTATS; i++) { - if (i == MEM_CGROUP_STAT_SWAP && !do_swap_account) + for (i = 0; i < ARRAY_SIZE(memcg1_stats); i++) { + if (memcg1_stats[i] == MEMCG_SWAP && !do_swap_account) continue; - pr_cont(" %s:%luKB", mem_cgroup_stat_names[i], - K(mem_cgroup_read_stat(iter, i))); + pr_cont(" %s:%luKB", memcg1_stat_names[i], + K(mem_cgroup_read_stat(iter, memcg1_stats[i]))); } for (i = 0; i < NR_LRU_LISTS; i++) @@ -2362,7 +2369,7 @@ void mem_cgroup_split_huge_fixup(struct page *head) for (i = 1; i < HPAGE_PMD_NR; i++) head[i].mem_cgroup = head->mem_cgroup; - __this_cpu_sub(head->mem_cgroup->stat->count[MEM_CGROUP_STAT_RSS_HUGE], + __this_cpu_sub(head->mem_cgroup->stat->count[MEMCG_RSS_HUGE], HPAGE_PMD_NR); } #endif /* CONFIG_TRANSPARENT_HUGEPAGE */ @@ -2372,7 +2379,7 @@ static void mem_cgroup_swap_statistics(struct mem_cgroup *memcg, bool charge) { int val = (charge) ? 1 : -1; - this_cpu_add(memcg->stat->count[MEM_CGROUP_STAT_SWAP], val); + this_cpu_add(memcg->stat->count[MEMCG_SWAP], val); } /** @@ -2731,13 +2738,10 @@ static unsigned long mem_cgroup_usage(struct mem_cgroup *memcg, bool swap) struct mem_cgroup *iter; for_each_mem_cgroup_tree(iter, memcg) { - val += mem_cgroup_read_stat(iter, - MEM_CGROUP_STAT_CACHE); - val += mem_cgroup_read_stat(iter, - MEM_CGROUP_STAT_RSS); + val += mem_cgroup_read_stat(iter, MEMCG_CACHE); + val += mem_cgroup_read_stat(iter, MEMCG_RSS); if (swap) - val += mem_cgroup_read_stat(iter, - MEM_CGROUP_STAT_SWAP); + val += mem_cgroup_read_stat(iter, MEMCG_SWAP); } } else { if (!swap) @@ -3134,15 +3138,15 @@ static int memcg_stat_show(struct seq_file *m, void *v) struct mem_cgroup *mi; unsigned int i; - BUILD_BUG_ON(ARRAY_SIZE(mem_cgroup_stat_names) != - MEM_CGROUP_STAT_NSTATS); + BUILD_BUG_ON(ARRAY_SIZE(memcg1_stat_names) != ARRAY_SIZE(memcg1_stats)); BUILD_BUG_ON(ARRAY_SIZE(mem_cgroup_lru_names) != NR_LRU_LISTS); - for (i = 0; i < MEM_CGROUP_STAT_NSTATS; i++) { - if (i == MEM_CGROUP_STAT_SWAP && !do_memsw_account()) + for (i = 0; i < ARRAY_SIZE(memcg1_stats); i++) { + if (memcg1_stats[i] == MEMCG_SWAP && !do_memsw_account()) continue; - seq_printf(m, "%s %lu\n", mem_cgroup_stat_names[i], - mem_cgroup_read_stat(memcg, i) * PAGE_SIZE); + seq_printf(m, "%s %lu\n", memcg1_stat_names[i], + mem_cgroup_read_stat(memcg, memcg1_stats[i]) * + PAGE_SIZE); } for (i = 0; i < ARRAY_SIZE(memcg1_events); i++) @@ -3165,14 +3169,15 @@ static int memcg_stat_show(struct seq_file *m, void *v) seq_printf(m, "hierarchical_memsw_limit %llu\n", (u64)memsw * PAGE_SIZE); - for (i = 0; i < MEM_CGROUP_STAT_NSTATS; i++) { + for (i = 0; i < ARRAY_SIZE(memcg1_stats); i++) { unsigned long long val = 0; - if (i == MEM_CGROUP_STAT_SWAP && !do_memsw_account()) + if (memcg1_stats[i] == MEMCG_SWAP && !do_memsw_account()) continue; for_each_mem_cgroup_tree(mi, memcg) - val += mem_cgroup_read_stat(mi, i) * PAGE_SIZE; - seq_printf(m, "total_%s %llu\n", mem_cgroup_stat_names[i], val); + val += mem_cgroup_read_stat(mi, memcg1_stats[i]) * + PAGE_SIZE; + seq_printf(m, "total_%s %llu\n", memcg1_stat_names[i], val); } for (i = 0; i < ARRAY_SIZE(memcg1_events); i++) { @@ -3645,10 +3650,10 @@ void mem_cgroup_wb_stats(struct bdi_writeback *wb, unsigned long *pfilepages, struct mem_cgroup *memcg = mem_cgroup_from_css(wb->memcg_css); struct mem_cgroup *parent; - *pdirty = mem_cgroup_read_stat(memcg, MEM_CGROUP_STAT_DIRTY); + *pdirty = mem_cgroup_read_stat(memcg, NR_FILE_DIRTY); /* this should eventually include NR_UNSTABLE_NFS */ - *pwriteback = mem_cgroup_read_stat(memcg, MEM_CGROUP_STAT_WRITEBACK); + *pwriteback = mem_cgroup_read_stat(memcg, NR_WRITEBACK); *pfilepages = mem_cgroup_nr_lru_pages(memcg, (1 << LRU_INACTIVE_FILE) | (1 << LRU_ACTIVE_FILE)); *pheadroom = PAGE_COUNTER_MAX; @@ -4504,10 +4509,8 @@ static int mem_cgroup_move_account(struct page *page, spin_lock_irqsave(&from->move_lock, flags); if (!anon && page_mapped(page)) { - __this_cpu_sub(from->stat->count[MEM_CGROUP_STAT_FILE_MAPPED], - nr_pages); - __this_cpu_add(to->stat->count[MEM_CGROUP_STAT_FILE_MAPPED], - nr_pages); + __this_cpu_sub(from->stat->count[NR_FILE_MAPPED], nr_pages); + __this_cpu_add(to->stat->count[NR_FILE_MAPPED], nr_pages); } /* @@ -4519,18 +4522,16 @@ static int mem_cgroup_move_account(struct page *page, struct address_space *mapping = page_mapping(page); if (mapping_cap_account_dirty(mapping)) { - __this_cpu_sub(from->stat->count[MEM_CGROUP_STAT_DIRTY], + __this_cpu_sub(from->stat->count[NR_FILE_DIRTY], nr_pages); - __this_cpu_add(to->stat->count[MEM_CGROUP_STAT_DIRTY], + __this_cpu_add(to->stat->count[NR_FILE_DIRTY], nr_pages); } } if (PageWriteback(page)) { - __this_cpu_sub(from->stat->count[MEM_CGROUP_STAT_WRITEBACK], - nr_pages); - __this_cpu_add(to->stat->count[MEM_CGROUP_STAT_WRITEBACK], - nr_pages); + __this_cpu_sub(from->stat->count[NR_WRITEBACK], nr_pages); + __this_cpu_add(to->stat->count[NR_WRITEBACK], nr_pages); } /* @@ -5190,9 +5191,9 @@ static int memory_stat_show(struct seq_file *m, void *v) tree_events(memcg, events); seq_printf(m, "anon %llu\n", - (u64)stat[MEM_CGROUP_STAT_RSS] * PAGE_SIZE); + (u64)stat[MEMCG_RSS] * PAGE_SIZE); seq_printf(m, "file %llu\n", - (u64)stat[MEM_CGROUP_STAT_CACHE] * PAGE_SIZE); + (u64)stat[MEMCG_CACHE] * PAGE_SIZE); seq_printf(m, "kernel_stack %llu\n", (u64)stat[MEMCG_KERNEL_STACK_KB] * 1024); seq_printf(m, "slab %llu\n", @@ -5202,13 +5203,13 @@ static int memory_stat_show(struct seq_file *m, void *v) (u64)stat[MEMCG_SOCK] * PAGE_SIZE); seq_printf(m, "shmem %llu\n", - (u64)stat[MEM_CGROUP_STAT_SHMEM] * PAGE_SIZE); + (u64)stat[NR_SHMEM] * PAGE_SIZE); seq_printf(m, "file_mapped %llu\n", - (u64)stat[MEM_CGROUP_STAT_FILE_MAPPED] * PAGE_SIZE); + (u64)stat[NR_FILE_MAPPED] * PAGE_SIZE); seq_printf(m, "file_dirty %llu\n", - (u64)stat[MEM_CGROUP_STAT_DIRTY] * PAGE_SIZE); + (u64)stat[NR_FILE_DIRTY] * PAGE_SIZE); seq_printf(m, "file_writeback %llu\n", - (u64)stat[MEM_CGROUP_STAT_WRITEBACK] * PAGE_SIZE); + (u64)stat[NR_WRITEBACK] * PAGE_SIZE); for (i = 0; i < NR_LRU_LISTS; i++) { struct mem_cgroup *mi; @@ -5231,11 +5232,11 @@ static int memory_stat_show(struct seq_file *m, void *v) seq_printf(m, "pgmajfault %lu\n", events[PGMAJFAULT]); seq_printf(m, "workingset_refault %lu\n", - stat[MEMCG_WORKINGSET_REFAULT]); + stat[WORKINGSET_REFAULT]); seq_printf(m, "workingset_activate %lu\n", - stat[MEMCG_WORKINGSET_ACTIVATE]); + stat[WORKINGSET_ACTIVATE]); seq_printf(m, "workingset_nodereclaim %lu\n", - stat[MEMCG_WORKINGSET_NODERECLAIM]); + stat[WORKINGSET_NODERECLAIM]); return 0; } @@ -5492,10 +5493,10 @@ static void uncharge_batch(struct mem_cgroup *memcg, unsigned long pgpgout, } local_irq_save(flags); - __this_cpu_sub(memcg->stat->count[MEM_CGROUP_STAT_RSS], nr_anon); - __this_cpu_sub(memcg->stat->count[MEM_CGROUP_STAT_CACHE], nr_file); - __this_cpu_sub(memcg->stat->count[MEM_CGROUP_STAT_RSS_HUGE], nr_huge); - __this_cpu_sub(memcg->stat->count[MEM_CGROUP_STAT_SHMEM], nr_shmem); + __this_cpu_sub(memcg->stat->count[MEMCG_RSS], nr_anon); + __this_cpu_sub(memcg->stat->count[MEMCG_CACHE], nr_file); + __this_cpu_sub(memcg->stat->count[MEMCG_RSS_HUGE], nr_huge); + __this_cpu_sub(memcg->stat->count[NR_SHMEM], nr_shmem); __this_cpu_add(memcg->stat->events[PGPGOUT], pgpgout); __this_cpu_add(memcg->stat->nr_page_events, nr_pages); memcg_check_events(memcg, dummy_page); diff --git a/mm/page-writeback.c b/mm/page-writeback.c index 33df0583edb9..777711203809 100644 --- a/mm/page-writeback.c +++ b/mm/page-writeback.c @@ -2427,7 +2427,7 @@ void account_page_dirtied(struct page *page, struct address_space *mapping) inode_attach_wb(inode, page); wb = inode_to_wb(inode); - mem_cgroup_inc_page_stat(page, MEM_CGROUP_STAT_DIRTY); + mem_cgroup_inc_page_stat(page, NR_FILE_DIRTY); __inc_node_page_state(page, NR_FILE_DIRTY); __inc_zone_page_state(page, NR_ZONE_WRITE_PENDING); __inc_node_page_state(page, NR_DIRTIED); @@ -2449,7 +2449,7 @@ void account_page_cleaned(struct page *page, struct address_space *mapping, struct bdi_writeback *wb) { if (mapping_cap_account_dirty(mapping)) { - mem_cgroup_dec_page_stat(page, MEM_CGROUP_STAT_DIRTY); + mem_cgroup_dec_page_stat(page, NR_FILE_DIRTY); dec_node_page_state(page, NR_FILE_DIRTY); dec_zone_page_state(page, NR_ZONE_WRITE_PENDING); dec_wb_stat(wb, WB_RECLAIMABLE); @@ -2706,7 +2706,7 @@ int clear_page_dirty_for_io(struct page *page) */ wb = unlocked_inode_to_wb_begin(inode, &locked); if (TestClearPageDirty(page)) { - mem_cgroup_dec_page_stat(page, MEM_CGROUP_STAT_DIRTY); + mem_cgroup_dec_page_stat(page, NR_FILE_DIRTY); dec_node_page_state(page, NR_FILE_DIRTY); dec_zone_page_state(page, NR_ZONE_WRITE_PENDING); dec_wb_stat(wb, WB_RECLAIMABLE); @@ -2753,7 +2753,7 @@ int test_clear_page_writeback(struct page *page) ret = TestClearPageWriteback(page); } if (ret) { - mem_cgroup_dec_page_stat(page, MEM_CGROUP_STAT_WRITEBACK); + mem_cgroup_dec_page_stat(page, NR_WRITEBACK); dec_node_page_state(page, NR_WRITEBACK); dec_zone_page_state(page, NR_ZONE_WRITE_PENDING); inc_node_page_state(page, NR_WRITTEN); @@ -2808,7 +2808,7 @@ int __test_set_page_writeback(struct page *page, bool keep_write) ret = TestSetPageWriteback(page); } if (!ret) { - mem_cgroup_inc_page_stat(page, MEM_CGROUP_STAT_WRITEBACK); + mem_cgroup_inc_page_stat(page, NR_WRITEBACK); inc_node_page_state(page, NR_WRITEBACK); inc_zone_page_state(page, NR_ZONE_WRITE_PENDING); } diff --git a/mm/rmap.c b/mm/rmap.c index e303fdbee561..a6d018c4a13a 100644 --- a/mm/rmap.c +++ b/mm/rmap.c @@ -1158,7 +1158,7 @@ void page_add_file_rmap(struct page *page, bool compound) goto out; } __mod_node_page_state(page_pgdat(page), NR_FILE_MAPPED, nr); - mem_cgroup_update_page_stat(page, MEM_CGROUP_STAT_FILE_MAPPED, nr); + mem_cgroup_update_page_stat(page, NR_FILE_MAPPED, nr); out: unlock_page_memcg(page); } @@ -1198,7 +1198,7 @@ static void page_remove_file_rmap(struct page *page, bool compound) * pte lock(a spinlock) is held, which implies preemption disabled. */ __mod_node_page_state(page_pgdat(page), NR_FILE_MAPPED, -nr); - mem_cgroup_update_page_stat(page, MEM_CGROUP_STAT_FILE_MAPPED, -nr); + mem_cgroup_update_page_stat(page, NR_FILE_MAPPED, -nr); if (unlikely(PageMlocked(page))) clear_page_mlock(page); diff --git a/mm/vmscan.c b/mm/vmscan.c index fbec74af2b69..417b6657e994 100644 --- a/mm/vmscan.c +++ b/mm/vmscan.c @@ -2046,8 +2046,7 @@ static bool inactive_list_is_low(struct lruvec *lruvec, bool file, active = lruvec_lru_size(lruvec, active_lru, sc->reclaim_idx); if (memcg) - refaults = mem_cgroup_read_stat(memcg, - MEMCG_WORKINGSET_ACTIVATE); + refaults = mem_cgroup_read_stat(memcg, WORKINGSET_ACTIVATE); else refaults = node_page_state(pgdat, WORKINGSET_ACTIVATE); @@ -2735,7 +2734,7 @@ static void snapshot_refaults(struct mem_cgroup *root_memcg, pg_data_t *pgdat) if (memcg) refaults = mem_cgroup_read_stat(memcg, - MEMCG_WORKINGSET_ACTIVATE); + WORKINGSET_ACTIVATE); else refaults = node_page_state(pgdat, WORKINGSET_ACTIVATE); diff --git a/mm/workingset.c b/mm/workingset.c index 51c6f61d4cea..37fc1057cd86 100644 --- a/mm/workingset.c +++ b/mm/workingset.c @@ -289,11 +289,11 @@ bool workingset_refault(void *shadow) refault_distance = (refault - eviction) & EVICTION_MASK; inc_node_state(pgdat, WORKINGSET_REFAULT); - mem_cgroup_inc_stat(memcg, MEMCG_WORKINGSET_REFAULT); + mem_cgroup_inc_stat(memcg, WORKINGSET_REFAULT); if (refault_distance <= active_file) { inc_node_state(pgdat, WORKINGSET_ACTIVATE); - mem_cgroup_inc_stat(memcg, MEMCG_WORKINGSET_ACTIVATE); + mem_cgroup_inc_stat(memcg, WORKINGSET_ACTIVATE); rcu_read_unlock(); return true; } @@ -475,8 +475,7 @@ static enum lru_status shadow_lru_isolate(struct list_head *item, if (WARN_ON_ONCE(node->exceptional)) goto out_invalid; inc_node_state(page_pgdat(virt_to_page(node)), WORKINGSET_NODERECLAIM); - mem_cgroup_inc_page_stat(virt_to_page(node), - MEMCG_WORKINGSET_NODERECLAIM); + mem_cgroup_inc_page_stat(virt_to_page(node), WORKINGSET_NODERECLAIM); __radix_tree_delete_node(&mapping->page_tree, node, workingset_update_node, mapping); -- cgit v1.2.3 From ccda7f4360be86b87497c50d1f58aab3fd85a9a5 Mon Sep 17 00:00:00 2001 From: Johannes Weiner Date: Wed, 3 May 2017 14:55:16 -0700 Subject: mm: memcontrol: use node page state naming scheme for memcg The memory controllers stat function names are awkwardly long and arbitrarily different from the zone and node stat functions. The current interface is named: mem_cgroup_read_stat() mem_cgroup_update_stat() mem_cgroup_inc_stat() mem_cgroup_dec_stat() mem_cgroup_update_page_stat() mem_cgroup_inc_page_stat() mem_cgroup_dec_page_stat() This patch renames it to match the corresponding node stat functions: memcg_page_state() [node_page_state()] mod_memcg_state() [mod_node_state()] inc_memcg_state() [inc_node_state()] dec_memcg_state() [dec_node_state()] mod_memcg_page_state() [mod_node_page_state()] inc_memcg_page_state() [inc_node_page_state()] dec_memcg_page_state() [dec_node_page_state()] Link: http://lkml.kernel.org/r/20170404220148.28338-4-hannes@cmpxchg.org Signed-off-by: Johannes Weiner Acked-by: Vladimir Davydov Acked-by: Michal Hocko Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/memcontrol.h | 73 +++++++++++++++++++++++----------------------- mm/memcontrol.c | 38 ++++++++++++------------ mm/page-writeback.c | 10 +++---- mm/rmap.c | 4 +-- mm/vmscan.c | 5 ++-- mm/workingset.c | 6 ++-- 6 files changed, 68 insertions(+), 68 deletions(-) (limited to 'include/linux/memcontrol.h') diff --git a/include/linux/memcontrol.h b/include/linux/memcontrol.h index 0fa1f5de6841..899949bbb2f9 100644 --- a/include/linux/memcontrol.h +++ b/include/linux/memcontrol.h @@ -472,8 +472,8 @@ extern int do_swap_account; void lock_page_memcg(struct page *page); void unlock_page_memcg(struct page *page); -static inline unsigned long mem_cgroup_read_stat(struct mem_cgroup *memcg, - enum memcg_stat_item idx) +static inline unsigned long memcg_page_state(struct mem_cgroup *memcg, + enum memcg_stat_item idx) { long val = 0; int cpu; @@ -487,27 +487,27 @@ static inline unsigned long mem_cgroup_read_stat(struct mem_cgroup *memcg, return val; } -static inline void mem_cgroup_update_stat(struct mem_cgroup *memcg, - enum memcg_stat_item idx, int val) +static inline void mod_memcg_state(struct mem_cgroup *memcg, + enum memcg_stat_item idx, int val) { if (!mem_cgroup_disabled()) this_cpu_add(memcg->stat->count[idx], val); } -static inline void mem_cgroup_inc_stat(struct mem_cgroup *memcg, - enum memcg_stat_item idx) +static inline void inc_memcg_state(struct mem_cgroup *memcg, + enum memcg_stat_item idx) { - mem_cgroup_update_stat(memcg, idx, 1); + mod_memcg_state(memcg, idx, 1); } -static inline void mem_cgroup_dec_stat(struct mem_cgroup *memcg, - enum memcg_stat_item idx) +static inline void dec_memcg_state(struct mem_cgroup *memcg, + enum memcg_stat_item idx) { - mem_cgroup_update_stat(memcg, idx, -1); + mod_memcg_state(memcg, idx, -1); } /** - * mem_cgroup_update_page_stat - update page state statistics + * mod_memcg_page_state - update page state statistics * @page: the page * @idx: page state item to account * @val: number of pages (positive or negative) @@ -518,28 +518,28 @@ static inline void mem_cgroup_dec_stat(struct mem_cgroup *memcg, * * lock_page(page) or lock_page_memcg(page) * if (TestClearPageState(page)) - * mem_cgroup_update_page_stat(page, state, -1); + * mod_memcg_page_state(page, state, -1); * unlock_page(page) or unlock_page_memcg(page) * * Kernel pages are an exception to this, since they'll never move. */ -static inline void mem_cgroup_update_page_stat(struct page *page, - enum memcg_stat_item idx, int val) +static inline void mod_memcg_page_state(struct page *page, + enum memcg_stat_item idx, int val) { if (page->mem_cgroup) - mem_cgroup_update_stat(page->mem_cgroup, idx, val); + mod_memcg_state(page->mem_cgroup, idx, val); } -static inline void mem_cgroup_inc_page_stat(struct page *page, - enum memcg_stat_item idx) +static inline void inc_memcg_page_state(struct page *page, + enum memcg_stat_item idx) { - mem_cgroup_update_page_stat(page, idx, 1); + mod_memcg_page_state(page, idx, 1); } -static inline void mem_cgroup_dec_page_stat(struct page *page, - enum memcg_stat_item idx) +static inline void dec_memcg_page_state(struct page *page, + enum memcg_stat_item idx) { - mem_cgroup_update_page_stat(page, idx, -1); + mod_memcg_page_state(page, idx, -1); } unsigned long mem_cgroup_soft_limit_reclaim(pg_data_t *pgdat, int order, @@ -739,40 +739,41 @@ static inline bool mem_cgroup_oom_synchronize(bool wait) return false; } -static inline unsigned long mem_cgroup_read_stat(struct mem_cgroup *memcg, - enum mem_cgroup_stat_index idx) +static inline unsigned long memcg_page_state(struct mem_cgroup *memcg, + enum memcg_stat_item idx) { return 0; } -static inline void mem_cgroup_update_stat(struct mem_cgroup *memcg, - enum memcg_stat_item idx, int val) +static inline void mod_memcg_state(struct mem_cgroup *memcg, + enum memcg_stat_item idx, + int nr) { } -static inline void mem_cgroup_inc_stat(struct mem_cgroup *memcg, - enum memcg_stat_item idx) +static inline void inc_memcg_state(struct mem_cgroup *memcg, + enum memcg_stat_item idx) { } -static inline void mem_cgroup_dec_stat(struct mem_cgroup *memcg, - enum memcg_stat_item idx) +static inline void dec_memcg_state(struct mem_cgroup *memcg, + enum memcg_stat_item idx) { } -static inline void mem_cgroup_update_page_stat(struct page *page, - enum memcg_stat_item idx, - int nr) +static inline void mod_memcg_page_state(struct page *page, + enum memcg_stat_item idx, + int nr) { } -static inline void mem_cgroup_inc_page_stat(struct page *page, - enum memcg_stat_item idx) +static inline void inc_memcg_page_state(struct page *page, + enum memcg_stat_item idx) { } -static inline void mem_cgroup_dec_page_stat(struct page *page, - enum memcg_stat_item idx) +static inline void dec_memcg_page_state(struct page *page, + enum memcg_stat_item idx) { } diff --git a/mm/memcontrol.c b/mm/memcontrol.c index 6fe4c7fafbfc..ff73899af61a 100644 --- a/mm/memcontrol.c +++ b/mm/memcontrol.c @@ -552,8 +552,8 @@ mem_cgroup_largest_soft_limit_node(struct mem_cgroup_tree_per_node *mctz) * implemented. */ -static unsigned long mem_cgroup_read_events(struct mem_cgroup *memcg, - enum memcg_event_item event) +static unsigned long memcg_sum_events(struct mem_cgroup *memcg, + enum memcg_event_item event) { unsigned long val = 0; int cpu; @@ -1180,7 +1180,7 @@ void mem_cgroup_print_oom_info(struct mem_cgroup *memcg, struct task_struct *p) if (memcg1_stats[i] == MEMCG_SWAP && !do_swap_account) continue; pr_cont(" %s:%luKB", memcg1_stat_names[i], - K(mem_cgroup_read_stat(iter, memcg1_stats[i]))); + K(memcg_page_state(iter, memcg1_stats[i]))); } for (i = 0; i < NR_LRU_LISTS; i++) @@ -2713,7 +2713,7 @@ static void tree_stat(struct mem_cgroup *memcg, unsigned long *stat) for_each_mem_cgroup_tree(iter, memcg) { for (i = 0; i < MEMCG_NR_STAT; i++) - stat[i] += mem_cgroup_read_stat(iter, i); + stat[i] += memcg_page_state(iter, i); } } @@ -2726,7 +2726,7 @@ static void tree_events(struct mem_cgroup *memcg, unsigned long *events) for_each_mem_cgroup_tree(iter, memcg) { for (i = 0; i < MEMCG_NR_EVENTS; i++) - events[i] += mem_cgroup_read_events(iter, i); + events[i] += memcg_sum_events(iter, i); } } @@ -2738,10 +2738,10 @@ static unsigned long mem_cgroup_usage(struct mem_cgroup *memcg, bool swap) struct mem_cgroup *iter; for_each_mem_cgroup_tree(iter, memcg) { - val += mem_cgroup_read_stat(iter, MEMCG_CACHE); - val += mem_cgroup_read_stat(iter, MEMCG_RSS); + val += memcg_page_state(iter, MEMCG_CACHE); + val += memcg_page_state(iter, MEMCG_RSS); if (swap) - val += mem_cgroup_read_stat(iter, MEMCG_SWAP); + val += memcg_page_state(iter, MEMCG_SWAP); } } else { if (!swap) @@ -3145,13 +3145,13 @@ static int memcg_stat_show(struct seq_file *m, void *v) if (memcg1_stats[i] == MEMCG_SWAP && !do_memsw_account()) continue; seq_printf(m, "%s %lu\n", memcg1_stat_names[i], - mem_cgroup_read_stat(memcg, memcg1_stats[i]) * + memcg_page_state(memcg, memcg1_stats[i]) * PAGE_SIZE); } for (i = 0; i < ARRAY_SIZE(memcg1_events); i++) seq_printf(m, "%s %lu\n", memcg1_event_names[i], - mem_cgroup_read_events(memcg, memcg1_events[i])); + memcg_sum_events(memcg, memcg1_events[i])); for (i = 0; i < NR_LRU_LISTS; i++) seq_printf(m, "%s %lu\n", mem_cgroup_lru_names[i], @@ -3175,7 +3175,7 @@ static int memcg_stat_show(struct seq_file *m, void *v) if (memcg1_stats[i] == MEMCG_SWAP && !do_memsw_account()) continue; for_each_mem_cgroup_tree(mi, memcg) - val += mem_cgroup_read_stat(mi, memcg1_stats[i]) * + val += memcg_page_state(mi, memcg1_stats[i]) * PAGE_SIZE; seq_printf(m, "total_%s %llu\n", memcg1_stat_names[i], val); } @@ -3184,7 +3184,7 @@ static int memcg_stat_show(struct seq_file *m, void *v) unsigned long long val = 0; for_each_mem_cgroup_tree(mi, memcg) - val += mem_cgroup_read_events(mi, memcg1_events[i]); + val += memcg_sum_events(mi, memcg1_events[i]); seq_printf(m, "total_%s %llu\n", memcg1_event_names[i], val); } @@ -3650,10 +3650,10 @@ void mem_cgroup_wb_stats(struct bdi_writeback *wb, unsigned long *pfilepages, struct mem_cgroup *memcg = mem_cgroup_from_css(wb->memcg_css); struct mem_cgroup *parent; - *pdirty = mem_cgroup_read_stat(memcg, NR_FILE_DIRTY); + *pdirty = memcg_page_state(memcg, NR_FILE_DIRTY); /* this should eventually include NR_UNSTABLE_NFS */ - *pwriteback = mem_cgroup_read_stat(memcg, NR_WRITEBACK); + *pwriteback = memcg_page_state(memcg, NR_WRITEBACK); *pfilepages = mem_cgroup_nr_lru_pages(memcg, (1 << LRU_INACTIVE_FILE) | (1 << LRU_ACTIVE_FILE)); *pheadroom = PAGE_COUNTER_MAX; @@ -4515,7 +4515,7 @@ static int mem_cgroup_move_account(struct page *page, /* * move_lock grabbed above and caller set from->moving_account, so - * mem_cgroup_update_page_stat() will serialize updates to PageDirty. + * mod_memcg_page_state will serialize updates to PageDirty. * So mapping should be stable for dirty pages. */ if (!anon && PageDirty(page)) { @@ -5161,10 +5161,10 @@ static int memory_events_show(struct seq_file *m, void *v) { struct mem_cgroup *memcg = mem_cgroup_from_css(seq_css(m)); - seq_printf(m, "low %lu\n", mem_cgroup_read_events(memcg, MEMCG_LOW)); - seq_printf(m, "high %lu\n", mem_cgroup_read_events(memcg, MEMCG_HIGH)); - seq_printf(m, "max %lu\n", mem_cgroup_read_events(memcg, MEMCG_MAX)); - seq_printf(m, "oom %lu\n", mem_cgroup_read_events(memcg, MEMCG_OOM)); + seq_printf(m, "low %lu\n", memcg_sum_events(memcg, MEMCG_LOW)); + seq_printf(m, "high %lu\n", memcg_sum_events(memcg, MEMCG_HIGH)); + seq_printf(m, "max %lu\n", memcg_sum_events(memcg, MEMCG_MAX)); + seq_printf(m, "oom %lu\n", memcg_sum_events(memcg, MEMCG_OOM)); return 0; } diff --git a/mm/page-writeback.c b/mm/page-writeback.c index 777711203809..2359608d2568 100644 --- a/mm/page-writeback.c +++ b/mm/page-writeback.c @@ -2427,7 +2427,7 @@ void account_page_dirtied(struct page *page, struct address_space *mapping) inode_attach_wb(inode, page); wb = inode_to_wb(inode); - mem_cgroup_inc_page_stat(page, NR_FILE_DIRTY); + inc_memcg_page_state(page, NR_FILE_DIRTY); __inc_node_page_state(page, NR_FILE_DIRTY); __inc_zone_page_state(page, NR_ZONE_WRITE_PENDING); __inc_node_page_state(page, NR_DIRTIED); @@ -2449,7 +2449,7 @@ void account_page_cleaned(struct page *page, struct address_space *mapping, struct bdi_writeback *wb) { if (mapping_cap_account_dirty(mapping)) { - mem_cgroup_dec_page_stat(page, NR_FILE_DIRTY); + dec_memcg_page_state(page, NR_FILE_DIRTY); dec_node_page_state(page, NR_FILE_DIRTY); dec_zone_page_state(page, NR_ZONE_WRITE_PENDING); dec_wb_stat(wb, WB_RECLAIMABLE); @@ -2706,7 +2706,7 @@ int clear_page_dirty_for_io(struct page *page) */ wb = unlocked_inode_to_wb_begin(inode, &locked); if (TestClearPageDirty(page)) { - mem_cgroup_dec_page_stat(page, NR_FILE_DIRTY); + dec_memcg_page_state(page, NR_FILE_DIRTY); dec_node_page_state(page, NR_FILE_DIRTY); dec_zone_page_state(page, NR_ZONE_WRITE_PENDING); dec_wb_stat(wb, WB_RECLAIMABLE); @@ -2753,7 +2753,7 @@ int test_clear_page_writeback(struct page *page) ret = TestClearPageWriteback(page); } if (ret) { - mem_cgroup_dec_page_stat(page, NR_WRITEBACK); + dec_memcg_page_state(page, NR_WRITEBACK); dec_node_page_state(page, NR_WRITEBACK); dec_zone_page_state(page, NR_ZONE_WRITE_PENDING); inc_node_page_state(page, NR_WRITTEN); @@ -2808,7 +2808,7 @@ int __test_set_page_writeback(struct page *page, bool keep_write) ret = TestSetPageWriteback(page); } if (!ret) { - mem_cgroup_inc_page_stat(page, NR_WRITEBACK); + inc_memcg_page_state(page, NR_WRITEBACK); inc_node_page_state(page, NR_WRITEBACK); inc_zone_page_state(page, NR_ZONE_WRITE_PENDING); } diff --git a/mm/rmap.c b/mm/rmap.c index a6d018c4a13a..3ff241f714eb 100644 --- a/mm/rmap.c +++ b/mm/rmap.c @@ -1158,7 +1158,7 @@ void page_add_file_rmap(struct page *page, bool compound) goto out; } __mod_node_page_state(page_pgdat(page), NR_FILE_MAPPED, nr); - mem_cgroup_update_page_stat(page, NR_FILE_MAPPED, nr); + mod_memcg_page_state(page, NR_FILE_MAPPED, nr); out: unlock_page_memcg(page); } @@ -1198,7 +1198,7 @@ static void page_remove_file_rmap(struct page *page, bool compound) * pte lock(a spinlock) is held, which implies preemption disabled. */ __mod_node_page_state(page_pgdat(page), NR_FILE_MAPPED, -nr); - mem_cgroup_update_page_stat(page, NR_FILE_MAPPED, -nr); + mod_memcg_page_state(page, NR_FILE_MAPPED, -nr); if (unlikely(PageMlocked(page))) clear_page_mlock(page); diff --git a/mm/vmscan.c b/mm/vmscan.c index 417b6657e994..4e7ed65842af 100644 --- a/mm/vmscan.c +++ b/mm/vmscan.c @@ -2046,7 +2046,7 @@ static bool inactive_list_is_low(struct lruvec *lruvec, bool file, active = lruvec_lru_size(lruvec, active_lru, sc->reclaim_idx); if (memcg) - refaults = mem_cgroup_read_stat(memcg, WORKINGSET_ACTIVATE); + refaults = memcg_page_state(memcg, WORKINGSET_ACTIVATE); else refaults = node_page_state(pgdat, WORKINGSET_ACTIVATE); @@ -2733,8 +2733,7 @@ static void snapshot_refaults(struct mem_cgroup *root_memcg, pg_data_t *pgdat) struct lruvec *lruvec; if (memcg) - refaults = mem_cgroup_read_stat(memcg, - WORKINGSET_ACTIVATE); + refaults = memcg_page_state(memcg, WORKINGSET_ACTIVATE); else refaults = node_page_state(pgdat, WORKINGSET_ACTIVATE); diff --git a/mm/workingset.c b/mm/workingset.c index 37fc1057cd86..b8c9ab678479 100644 --- a/mm/workingset.c +++ b/mm/workingset.c @@ -289,11 +289,11 @@ bool workingset_refault(void *shadow) refault_distance = (refault - eviction) & EVICTION_MASK; inc_node_state(pgdat, WORKINGSET_REFAULT); - mem_cgroup_inc_stat(memcg, WORKINGSET_REFAULT); + inc_memcg_state(memcg, WORKINGSET_REFAULT); if (refault_distance <= active_file) { inc_node_state(pgdat, WORKINGSET_ACTIVATE); - mem_cgroup_inc_stat(memcg, WORKINGSET_ACTIVATE); + inc_memcg_state(memcg, WORKINGSET_ACTIVATE); rcu_read_unlock(); return true; } @@ -475,7 +475,7 @@ static enum lru_status shadow_lru_isolate(struct list_head *item, if (WARN_ON_ONCE(node->exceptional)) goto out_invalid; inc_node_state(page_pgdat(virt_to_page(node)), WORKINGSET_NODERECLAIM); - mem_cgroup_inc_page_stat(virt_to_page(node), WORKINGSET_NODERECLAIM); + inc_memcg_page_state(virt_to_page(node), WORKINGSET_NODERECLAIM); __radix_tree_delete_node(&mapping->page_tree, node, workingset_update_node, mapping); -- cgit v1.2.3